130
Journal of Chemml and Engineering Data, Vol 24, No 2, 1979
Excess Thermodynamic Functions for Ternary Systems. 4. Total-Pressure Data and GE for Acetonitrile-Ethanol-Water at 50 "C Stephen R. Wilson, Rohit B. Patel, Michael M. Abbott," and Hendrick C. Van Ness* Chemical and Environmental Engineering Department, Rensselaer Polytechnic Institute, Troy, New York 12 18 1
Isothermal P - X data for the ternary system acetonitrile-ethanol-water at 50 OC are reported. Data are also presented for the constituent binaries at 50 OC. Reduction of the data by Barker's method allows calculation of GE. The data set reported here comprises VLE measurements for the system acetonitrile (1)-ethanol (2)-water (3) at 50 O C . Experimental results are presented for the three constituent binaries and for six runs on ternary mixtures formed by additions of each pure constituent to mixtures of the other two in molar proportions of 2 to 1. Data were taken with the total-pressure apparatus of Gibbs and Van Ness ( 7 ) , modified as described briefly in part 3 ( 5 ) of this series of papers. The acetonitrile was chromatoquality reagent from Matheson Coleman and Bell; the reagent-grade ethanol was supplied by U.S. Industrial Chemicals, and the water was doubly deionized. Except for degassing, all reagents were used as received, with indicated purities of at least 99.8 mol % . Vapor pressures of the pure constituents as measured during the course of these experiments and their comparison with literature values are reported in Table I. Because of the excellent reproducibility of these measurements, Pisatvalues in all calculations are fixed at average values.
Table 111. Total-Pressure Data for Acetonitrile (I)-Water (3) at 50 "C
ethanol (2)
water (3)
29.490 29.491 29.494 29.496 29.493 29.481 (11) 29.494 ( 4 ) 29.507 (10)
12.363 12.342 12.348 12.349 12.350 12.345 (3, 11) 12.349 ( 9 )
av value lit. values
0.0 0.0328 0.0974 0.1699 0.2261 0.2846 0.3556 0.42 15 0.5041 0.5748 0.6687 0.7418 0.8330 0.9026 0.9472 1 .0000
1.oooo 0.9672 0.9026 0.8301 0.77 39 0.7154 0.6444 0.5785 0.4959 0.4252 0.3313 0.2582 0.1670 0.0974 0.0528 0.0
12.348 22.829 32.279 35.358 36.180 36.672 37.009 37.262 37.527 37.777 38.020 38.146 37.91 1 37.149 36.146 33.860
x2
Table 1. Vapor Pressures of Pure Constituents at 50 " C in kPa
33.859 33.860 33.864 33.864 33.862 33.845 (8) 33.872 (6)
PlkPa
0.0 0.05 18 0.1418 0.2169 0.3163 0.4065 0.4864 0.5 642 0.6391 0.7063 0.7923 0.8724 0.9500 1.oooo
x2
1.0000 0.9482 0.8582 0.7831 0.6837 0.5 9 35 0.5 136 0.4358 0.3609 0.2937 0.2077 0.1276 0.0500 0.0
9.. E rl
PlkPa
29.496 33.343 38.000 39.878 41.203 41.952 42.313 42.395 42.239 41.851 40.854 39.204 36.499 33.864
0021-9568/79/1724-0130$01,00/0
12.363 16.247 18.5 18 20.686 22.344 23.393 24.303 25.005 25.514 26.258 27.030 27.658 28.495 29.121 29.474 29.518 29.490
reduction is by Barker's method, with analytical expressions for GE provided by Margules or modified Margules expressions for the binary systems and by the Wohl equation for the ternary data. The procedures are fully described elsewhere ( 1 , 2 ) . I n particular, the data for the acetonitrile (1)-ethanol (2) and the ethanol (2)-water (3) binaries are correlated by the Margules equation:
Table 11. Total-Pressure Data for Acetonitrile (I)-Ethanol (2) at 50 "C Xi
PlkPa
x3
1.0000 0.9675 0.9425 0.9080 0.8756 0.8413 0.7993 0.7612 0.7187 0.6402 0.5597 0.4786 0.3595 0.2299 0.1274 0.041 1 0.0
0.0 0.0 325 0.0575 0.0920 0.1244 0.1587 0.2007 0.2388 0.2813 0.3598 0.4403 0.5214 0.6405 0.7701 0.8726 0.9589 1.oooo
Tables I1 through I V give experimental values of total pressures for the three constituent binaries, and Table V contains all data for the six runs made with ternary mixtures. Data acetonitrile (1)
x3
Table IV. Total-Pressure Data for Ethanol (2)-Water (3) at 50 "C
Results and Correlations
present nark
XI
GE, - = [(Aj,xi RT
+ A,xj) - (Xjixi -I- X , ~ ~ ) X ~ ~ ] X + (1)X ,
Four parameters are required for acetonitrile-ethanol but only three for ethanol-water. I n the latter case
A,.rl = X..I1 For acetonitrile ( 1)-water (3) the five-parameter modified Margules equation is required for correlation of the data:
GE,3 gI3 I - = RT
-
I
(A31X1
+ A 13x3)
' I
cy 13cy3 1
0113x1
@ 1979 American Chemical Society
Ix 3
+ O3Ix3 + q x I x 3
J
x1x3
(2)
Journal of Chemical and Engineering Data, Vol. 24, No. 2, 1979 131 Table V. Total-Pressure Data for Acetonitrile (1)-Ethanol ( 2 ) - ~ a t e r(3) at 50 "C JC 2
Xl
x3
0.0 0.6674 0.6570 0.6327 0.6025 0.5695 0.5297 0.5 0 3 1 0.4699 0.4358 1.0000 0.3306 0.3233 0.3116 0.1969 0.2809 0. !613 0.1481 0.2319 0.1121 0.0
1.0000 0.3326 0.3273 0.3153 0.3002 0.2838 0.2639 0.2507 0.2341 0.2172 0.0 0.6694 0.6546 0.6310 0.6010 0.5687 0.5291 0.5023 0.4696 0.4295 1.oooo 0.6700 0.6543 0.6314 0.6036 0.5704 0.5 305 0.5040 0.4705 0.4378 0.0 0.3366 0.3296 0.3177 0.3022 0.2842 0.2646 0.25 11 0.2341 0.2174 0.0 0.0 0.021 1 0.0575 0.1056 0.1560 0.2186 0.2596 0.3103 0.3614 0.0 0.0 0.0203 0.0565 0.1028 0.1537 0.2 I37 0.2541 0.3161 0.3514
0.1 0. I234 0.1576 0. I991 0.1487 0.2082 0.2478 0.2977 0.3465 0.1
0.:I 0.0208 0.11561 0.1022 0.1559 0.2140 0.2544 0.3048 0.3544 1.0000 0.3433 0.3360 0.3235 0.3070 0.2897 0.2683 0.2542 0.2368 0.2193 0.'3 0.6697 0.6561 0.6318 0.6008 0.5667 0.5266 0.4995 0.15 80 0.1344
0.0 0.0 0.0157 0.0520 0.097 3 0.1467 0.2064 0.2462 0.2960 0.3470 0.0 0.0 0.0221 0.0574 0.1021 0.1504 0.2096 0.2496 0.2985 0.3584 0.0 0.3300 0.3223 0.3110 0.2973 0.2809 0.2613 0.2482 0.2318 0.2157
1.oooo 0.6634 0.6496 0.6262 0.5956 0.5599 0.5214 0.4945 0.46 1 1 0.4282 0.0 0.6567 0.6429 0.6190 0.5874 0.5543 0.5 131 0.4862 0.4529 0.4193 1.0000 0.3303 0.3236 0.31 17 0.2964 0.2796 0.2597 0.2464 0.2259 0.2 142
Table VI. Summary of Results for Binary Systems at 5 0 "C
acetonitrile (1)- acetonitrile (1)ethanol (2) water (3)
PlkPa 33.864 41.361 41.326 41.033 40.639 40.194 39.521 39.072 38.477 37.828 29.491 42.089 42.060 41.948 41.749 41.434 40.999 40.641 40.177 39.580 33.859 38.007 38.460 38.988 39.526 40.025 40.45 1 40.635 40.785 40.833 12.342 36.996 37.116 37.286 37.416 37.493 37.483 37.447 37.269 37.250 29.494 26.187 28.164 30.845 33.390 35.219 36.894 37.637 38.370 38.930 12.349 28.608 30.229 32.436 34.684 36.676 38.160 39.008 39.995 40.359
Pisat, kPa Pjsat, kPa V i L , crn3/rnol VjL, crn3/rnol Bii, crn3/mol Bjj, cm3/rnol Bij, cm3//mol Aij Aji
hij hji "ij "ji ri
=
g12
+ g13 + g23 + CxlxZx3
0.029 0.030 0.143 0.150
33.862 12.350 54.71 18.23 -4036 -1674 -281 2.587 ?- 0.007 2.000 t 0.009
1.202 f 0.071 0.578 I 0.044 -0.679 * 0.079 RMS AP,kPa 0.079 0.032 niax IAPI, kPa 0.176 0.063 XiaZ 0.5454 0.7560 P a z , kPa 42.419 38.140 a Pairs of components ;ire listed in the order i, I .
29.493 12.350 60.36 18.23 -1706 -1674 -948 1.697 ? 0.929 i 0.313 k 0.313 t
0.007 0.012 0.034 0.034
0.054 0.1 34 0.9439 29.5 37
Computer-generated contour diagram showing lines of constant P (in mmHg) on a liquid-mole-fractiongrid for acetonitrile (1)-ethanol (2)-water (3) at 50 O C . Figure 1.
parameters fixed at values as given in Table VI, yields for the ternary parameter the value
C = 3.686 & 0.007 The root-mean-square A f for the ternary data is 0.056 kPa; the maximum IAPI is 0.120 kPa. Inclusion of higher order ternary terms in eq 3 results in no significant improvement in the quality of correlation.
Discussion
The ternary data are well fit by the simple equation g123
33.862 29.493 54.71 60.36 -4036 -1706 -1142 1.292 * 1.235 ?0.047 ?0.615 t
ethanol (2)water (3)
(3)
Correlations for the goare provided by the binary data alone; the parameter Cis found by regression of just the ternary data. Second virial coefficients By required to account for vapor-phase nonidealities are estimated by the method of Tsonopoulos (72). Results of the correlations of data for the binary systems, together with all ancillary information, are summarized in Table VI. Correlation of the data for the ternary system, with binary
The ethanol-water binary is the only system for which literature values are available for comparison with our results. Pemberton and co-workers ( 10, 1 7 ) report total-pressure values at 50 OC as part of an extensive data set for both VLE and HE taken over a wide temperature range. Comparison of total pressures calculated from our correlation with their smoothed values shows only fair agreement:
RMS A f = 0.10 kPa max IAP I = 0.26 kPa
132 Journal of Chemical and Engineering Data, Vol. 24, No. 2, 1979 --7
-->
--i
--i
--i
--7
--i
--i
--i
4
--. --i
... --i
--i
--i
..,
--i
... 3
1
Figure 2. Computer-generated pictorial view of the P-x surface for acetonitrile (1)-ethanol (2)-water (3) at 50 O C . \2*
7%
,-,
1
Figure 4. Computer-generated pictorial view of the GE,,3-~surface for acetonitrile (1)-ethanol (2)-water (3) at 50 O C .
excess Gibbs function, liquid phase G ~RT/ total pressure azeotropic pressure vapor pressure of pure i universal gas constant absolute temperature molar volume of pure liquid i mole fraction, liquid phase azeotropic composition
GE
9 P P az pi sat R T Vi X Xaz
Greek Letters
a,,, CY/, 17
A,, A ,
A
parameters in eq 2 parameter in eq 2 parameters in eq 1 signifies a difference
Literature Cited Figure 3. Computer-generated contour diagram showing lines of constant GE,,3 (in J/rnol) on a liquid-mole-fraction grid for acetonitrile (1)-ethanol (2)-water (3) at 50 OC.
The results of this study are displayed pictorially by Figures 1 through 4. It is evident from Figure 1 that each binary system contains a maximum-pressure azeotrope. There is, however, no ternary azeotrope or other singular point on the pressurecomposition surface. Glossary
Aij,Aji Bf C
parameters in eq 1 and 2 second virial coefficient parameter in eq 3
Abbott. M. M.. Fioess. J. K.. Walsh, G. E., Van Ness, H. C., AIChE J . , 2 1; 72' ( 1975)'. Abbott, M. M., Van Ness, H. C., AIChE J . , 21, 62 (1975). Ambrose, D., Lawrenson, I. J., J . Chem. Thermodyn., 4, 755 (1972). Ambrose, D, Sprake, C. H. S.,Townsend, R., J . Chem. Thermodyn., 7, 185 (1975). DiElsi, D. P., Patel, R. B., Abbott, M. M., Van Ness, H. C., J . Chem. Eng. Data, 23, 242 (1978). Dreisbach, R. R., Adv. Chem. Ser., No. 29 (1967). Gibbs, R. E., Van Ness, H. C.. I r d Eng. Chem. Fundam., 11, 410 (1972). Heim, G., Bull. SOC. Chim. 8e/g., 42, 467 (1933). Keenan, J. H., Keyes, F. G., Hill, P. G., Moore, J. G., "Steam Tables", Wiley, New York, 1969. Larkin, J. A,, Pemberton, R. C., Natl. Phys. Lab. ( U . K . )Rep., 43 (1976). Pemberton. R. C.. Mash, C. J., J . Chem. Thermodyn., 10, 867 (1978). Tsonopouios, C., AIChE J . , 20, 263 (1974). Received fcf review August 4, 1978. Accepted January 8, 1979. Partial support (about 20%) for this work came from National Science Foundation Grant No. ENG75-17367 and the remainder from the Department of Chemical and Environmental Engineering, Rensselaer Polytechnic Institute.
